This change implements test cases to validate that:
scale up will be done in order
scale down in reverse order
if any of the pets will be unhealthy, scale up/down will halt
Change-Id: I4487a7c9823d94a2995bbb06accdfd8f3001354c
Automatic merge from submit-queue
improve readme for contributor experience
**What this PR does / why we need it**:
This improves the top-level README. Removes broken links, removes links to make you bounce to another page, adds proper links to docs.
Shuffles things around a bit and improves reading.
@kubernetes/contributor-experience
Automatic merge from submit-queue
fix leaking memory backed volumes of terminated pods
Currently, we allow volumes to remain mounted on the node, even though the pod is terminated. This creates a vector for a malicious user to exhaust memory on the node by creating memory backed volumes containing large files.
This PR removes memory backed volumes (emptyDir w/ medium Memory, secrets, configmaps) of terminated pods from the node.
@saad-ali @derekwaynecarr
Automatic merge from submit-queue
Adding statefulset to the list of things kubectl says it knows about
**What this PR does / why we need it**: Adding statefulset to the list of things kubectl says it knows about.
**Special notes for your reviewer**:
**Release note**:
<!-- Steps to write your release note:
1. Use the release-note-* labels to set the release note state (if you have access)
2. Enter your extended release note in the below block; leaving it blank means using the PR title as the release note. If no release note is required, just write `NONE`.
-->
```release-note
NONE
```
cc @kubernetes/sig-apps @erictune
Automatic merge from submit-queue
Bump fluentd-gcp image version
To propagate https://github.com/kubernetes/kubernetes/pull/36993
Including release noted for the whole bunch of fluentd changes (https://github.com/kubernetes/kubernetes/pull/35618 mainly):
```release-note
Default logging subsystem's resiliency was greatly improved, fluentd memory consumption and OOM error probability was reduced.
```
@piosz
Automatic merge from submit-queue
Node E2E: Remove kubelet related flags
This PR:
1. **Add `KubeletConfig` in test context.** The configuration is dynamically retrieved from kubelet `configz/` endpoint after test node registers itself.
2. **Change tests to look at `KubeletConfig` instead of other kubelet related fields in test context.**
3. **Remove kubelet related flags from node e2e.** Add `kubelet-flags` to pass kubelet flags all together.
Why do we need this change?
1. This is required by **node soaking test and node conformance test**.
* **For node soaking test.** We'll have one job start the test environment, and another job running the actual test again and again. It's hard to tell the test by flag how kubelet is configured.
* **For node conformance test.** Kubelet is usually started by user. It's hard and troublesome to tell the test by flag how kubelet is configured.
2. **No need to add another flag in node e2e for each kubelet flag we want to set in the test.** Just directly add the kubelet flag into `kubelet-flags`.
In the future,
1. Kubelet start logic should be moved outside of the test. We should have standard kubelet launcher or standard cloud-init node setup script etc.
2. The test should validate kubelet configuration:
* Whether the configuration is production ready?
* Whether the configuration has conflicts?
* Whether some configuration conflict with node e2e/conformance test is turned on (cloud provider etc.)? (Some configurations need to be turned off when running node e2e/conformance test)
* ...
/cc @kubernetes/sig-node
Automatic merge from submit-queue
Add e2e test for statefulset updates
Verify that one can (manually) update statefulset template
cc @erictune @foxish @kow3ns @kubernetes/sig-apps
Automatic merge from submit-queue
Wait for all Nodes to be schedulable before running e2e tests
This should fix the problem we're seeing when running tests on large clusters.
cc @dchen1107
Automatic merge from submit-queue
Delete taint annotation when removing last taint
It messes with debugging of tests failures.
cc @davidopp @kevin-wangzefeng
Automatic merge from submit-queue
Add more test cases to k8s e2e upgrade tests
**Special notes for your reviewer**:
Added guestbook, secrets, daemonsets, configmaps, jobs to e2e upgrade tests according to the discussions in #35078
Still need to run these test cases in real setup, raised a PR here for initial comments
@quinton-hoole
Automatic merge from submit-queue
Fix hostname truncate.
Fixes https://github.com/kubernetes/kubernetes/issues/36951.
This PR will keep truncating the hostname until the ending character is valid.
/cc @kubernetes/sig-node
Mark v1.5 because this is a bug fix.
/cc @saad-ali
Automatic merge from submit-queue
make kubectl create --edit iterate
`kubectl create --edit` is broken after #36148 merged.
`kubectl create --edit` will fail when a manifest that contains multiple resources.
I guess the root cause is that dynamic typer doesn't support a list of resources currently.
This PR makes `kubectl create --edit` iterate again as `kubectl create`.
Automatic merge from submit-queue
Add clarity/retries to proxy url test
Improve one segment of the kube-proxy networking test by:
1. Retrying for 30s
2. Bucketing into 2 failure modes
3. Adding some clarity by describing the exec pod on failure
Althought 1 shouldn't be necessary, I don't think we lose anything if the kube-proxy convenience endpoint doesn't respond immediately, and if it fails for 30s straight it is indicative of something that requires attention probably within 1.5.
Fixes https://github.com/kubernetes/kubernetes/issues/32436
Automatic merge from submit-queue
Require BOOTSTRAP_MIGRATION, JENKINS_BUILD_STARTED or JENKINS_BUILD_F…
If we do not do this then we upload finished.json and update latest-build.txt before the build fiinshes, breaking the SQ
Automatic merge from submit-queue
Enable NFSv4 and GlusterFS tests on cluster e2e tests
Enable NFSv4 and GlusterFS tests on cluster e2e tests for GCI images
only.
Automatic merge from submit-queue
Fix syntax error in local-up-cluster.sh when CGROUP_ROOT is set
Fix syntax error when `CGROUP_ROOT` is set, or it will complain a following error:
```
hack/local-up-cluster.sh: line 39: CGROUP_ROOT: "": syntax error: operand expected (error token is """")
```
cc/ @derekwaynecarr
Automatic merge from submit-queue
Handle Empty clusterCIDR
**What this PR does / why we need it**:
Handles empty clusterCIDR by skipping the corresponding rule.
**Which issue this PR fixes**
fixes#36652
**Special notes for your reviewer**:
1. Added test to check for presence/absence of XLB to SVC rule
2. Changed an error statement to log rules along with the error string in case of a failure; This ensures that full debug info is available in case of iptables-restore errors.
Empty clusterCIDR causes invalid rules generation.
Fixes issue #36652
Automatic merge from submit-queue
dockershim: remove sandbox container upon naming conflicts
We have observed that, after failing to create a container due to "device or
resource busy", docker may end up having inconsistent internal state. One
symptom is that docker will not report the existence of the "failed to create"
container, but if kubelet tries to create a new container with the same name,
docker will error out with a naming conflict message.
To work around this, this commit parses the creation error message and if there
is a naming conflict, it would attempt to remove the existing container.
